recall

A command for explicitly searching stored memories using a text query, with optional limits, categories, and minimum reliability scores. Memory retrieval is the process of finding earlier project knowledge relevant to the current question.

In plain words
What is it for?
For searching Memory Layer for topics such as authentication, database connections, naming rules, error handling, or other earlier project context.
Why use it?
It helps find past decisions, conventions, fixes, or patterns when automatic retrieval does not return what you need.

Command for Claude Code

Install

Getting it into your agent

One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.

agentmods
npx agentmods add commands/runtimenoteslabs/memory-layer/recall
Clone the repo
git clone --depth 1 https://github.com/runtimenoteslabs/memory-layer

Made for: Claude Code.

Per session 12 Only the description is in the session, so the agent can decide to use it. The body loads when it is invoked.
When invoked 430 The whole file, excluding the scripts and references it only reads on demand.
Security scan A 0 findings. Scan, not verified.
Origin original No closer match found in the catalogue.
Token cost

What it costs to keep this loaded

Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.

ModelPer sessionOnce invoked
Fable 5 $0.00012 $0.00430
Opus 5 $0.00006 $0.00215
Sonnet 5 $0.00002 $0.00086
Haiku 4.5 $0.00001 $0.00043

Measured 2d ago against content hash c2b1163d8a59, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

recall scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 2d ago.

A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.

Nothing flagged

None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.

.claude/commands/recall.md · 71 lines

What it actually says

Recall Command

Explicitly search the Memory Layer for relevant memories. Note that Agent Skills usually handle retrieval automatically based on conversation context, so this command is rarely needed.

Usage

/recall <query> [--limit N] [--category <cat>]

Options

Option Description Default
--limit N Maximum number of results 5
--category <cat> Filter by category all
--min-score <float> Minimum outcome score -1.0

Examples

# Search for authentication patterns
/recall "authentication patterns"

# Search with limit
/recall "database connection" --limit 10

# Search within a category
/recall "naming" --category convention

# Search for high-confidence memories only
/recall "error handling" --min-score 0.3

Implementation

Search memories and display results:

mem search "$QUERY" --format context --limit 5

When to Use

The memory-retrieval Agent Skill automatically retrieves memories when you:

  • Ask about past decisions ("what did we decide about...")
  • Reference conventions ("what's our convention for...")
  • Mention previous work ("last time we...", "we discussed...")

Use /recall explicitly when:

  • You want to browse all memories on a topic
  • You need more results than auto-retrieval provides
  • You want to filter by specific category or score
  • Auto-retrieval didn't surface what you were looking for

Result Format

Results are ranked by a hybrid score combining:

  • Semantic similarity (35%)
  • Outcome score (25%) - proven advice ranks higher
  • Recency (15%)
  • Frequency of use (15%)
  • Confidence (10%)

After reviewing results, use /outcome <id> worked|failed|partial to provide feedback.

Changes

What this file has done since we first saw it

Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.

  1. 2d ago First seen · 71 lines · 12 tokens per session scan A c2b1163d8a59

Subscribe to this mod's changes

recall is a command published in the GitHub repository runtimenoteslabs/memory-layer (10 stars, last pushed 3mo ago), licensed MIT. It adds 12 tokens to every session and 430 once invoked, about $0.0001 per session on Opus 5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-31.